diff options
author | elliott_c <ocielliottc@users.noreply.github.com> | 2003-12-12 14:37:23 +0000 |
---|---|---|
committer | elliott_c <ocielliottc@users.noreply.github.com> | 2003-12-12 14:37:23 +0000 |
commit | 49fda409cb50573f47244fb4beec20568b7cedf5 (patch) | |
tree | ecf29f01965485829b2b27c4c7b088e88ef1a55e | |
parent | 8a33be832cc26ffb94e0a0edb7dcc7ac216476dc (diff) | |
download | ATCD-49fda409cb50573f47244fb4beec20568b7cedf5.tar.gz |
ChangeLogTag: Fri Dec 12 08:37:04 2003 Chad Elliott <elliott_c@ociweb.com>
-rw-r--r-- | TAO/ChangeLog | 662 | ||||
-rw-r--r-- | TAO/orbsvcs/orbsvcs/PortableGroup/PG_Location_Hash.h | 1 |
2 files changed, 335 insertions, 328 deletions
diff --git a/TAO/ChangeLog b/TAO/ChangeLog index 8a6e886ceaa..eb135034086 100644 --- a/TAO/ChangeLog +++ b/TAO/ChangeLog @@ -1,6 +1,12 @@ -Fri Dec 12 09:46:33 UTC 2003 Johnny Willemsen <jwillemsen@remedy.nl> +Fri Dec 12 08:37:04 2003 Chad Elliott <elliott_c@ociweb.com> - * tests/RTScheduling/VoidData/VoidData.dsp: + * orbsvcs/orbsvcs/PortableGroup/PG_Location_Hash.h: + + Added #include of portablegroup_export.h. + +Fri Dec 12 09:46:33 UTC 2003 Johnny Willemsen <jwillemsen@remedy.nl> + + * tests/RTScheduling/VoidData/VoidData.dsp: Added missing libpaths Fri Dec 12 09:32:13 UTC 2003 Johnny Willemsen <jwillemsen@remedy.nl> @@ -10,49 +16,49 @@ Fri Dec 12 09:32:13 UTC 2003 Johnny Willemsen <jwillemsen@remedy.nl> Thu Dec 11 23:10:14 2003 Balachandran Natarajan <bala@dre.vanderbilt.edu> - * examples/OBV/Typed_Events/Typed_Events.mpc - * examples/POA/DSI/POA_DSI.mpc: - * examples/POA/Generic_Servant/POA_Generic_Servant.mpc: - * examples/POA/On_Demand_Loading/POA_On_Demand_Loading.mpc: - * examples/POA/TIE/POA_TIE.mpc: - * tests/POA/Excessive_Object_Deactivations/Excessive_Object_Deactivations.mpc: - * tests/RTCORBA/Explicit_Binding/RTCORBA_Explicit_Binding.mpc: - * tests/RTCORBA/Persistent_IOR/RTCORBA_Persistent_IOR.mpc: - * tests/RTCORBA/Profile_And_Endpoint_Selection/Profile_And_Endpoint_Selection.mpc: - * performance-tests/RTCorba/Multiple_Endpoints/Common/Common.mpc: - * orbsvcs/examples/ORT/ORT.mpc: + * examples/OBV/Typed_Events/Typed_Events.mpc + * examples/POA/DSI/POA_DSI.mpc: + * examples/POA/Generic_Servant/POA_Generic_Servant.mpc: + * examples/POA/On_Demand_Loading/POA_On_Demand_Loading.mpc: + * examples/POA/TIE/POA_TIE.mpc: + * tests/POA/Excessive_Object_Deactivations/Excessive_Object_Deactivations.mpc: + * tests/RTCORBA/Explicit_Binding/RTCORBA_Explicit_Binding.mpc: + * tests/RTCORBA/Persistent_IOR/RTCORBA_Persistent_IOR.mpc: + * tests/RTCORBA/Profile_And_Endpoint_Selection/Profile_And_Endpoint_Selection.mpc: + * performance-tests/RTCorba/Multiple_Endpoints/Common/Common.mpc: + * orbsvcs/examples/ORT/ORT.mpc: - Do not compile the above tests when minimum_corba is disabled. + Do not compile the above tests when minimum_corba is disabled. Thu Dec 11 22:22:36 2003 Balachandran Natarajan <bala@dre.vanderbilt.edu> - * tests/RTScheduling/VoidData/Makefile: + * tests/RTScheduling/VoidData/Makefile: - Added IDL compiler options. + Added IDL compiler options. - * tests/RTScheduling/VoidData/test.idl: + * tests/RTScheduling/VoidData/test.idl: - Cosmetic fix. + Cosmetic fix. - * examples/Kokyu_dsrt_schedulers/FP_Scheduling.idl: - * examples/Kokyu_dsrt_schedulers/MIF_Scheduling.idl: - * examples/Kokyu_dsrt_schedulers/MUF_Scheduling.idl: + * examples/Kokyu_dsrt_schedulers/FP_Scheduling.idl: + * examples/Kokyu_dsrt_schedulers/MIF_Scheduling.idl: + * examples/Kokyu_dsrt_schedulers/MUF_Scheduling.idl: - Added RTScheduler_include.idl instead of the regular file. + Added RTScheduler_include.idl instead of the regular file. Thu Dec 11 17:04:57 2003 Balachandran Natarajan <bala@dre.vanderbilt.edu> - * orbsvcs/performance-tests/RTEvent/lib/Loopback_Consumer.cpp: - * orbsvcs/performance-tests/RTEvent/lib/Loopback_Consumer.h: - * orbsvcs/performance-tests/RTEvent/lib/Send_Task.cpp: - * orbsvcs/performance-tests/RTEvent/lib/Send_Task.h: + * orbsvcs/performance-tests/RTEvent/lib/Loopback_Consumer.cpp: + * orbsvcs/performance-tests/RTEvent/lib/Loopback_Consumer.h: + * orbsvcs/performance-tests/RTEvent/lib/Send_Task.cpp: + * orbsvcs/performance-tests/RTEvent/lib/Send_Task.h: - Used the $TAO_ROOT/tao/Utils Servant_Var instead of the one in - the local directory. + Used the $TAO_ROOT/tao/Utils Servant_Var instead of the one in + the local directory. - * orbsvcs/performance-tests/RTEvent/lib/Makefile: + * orbsvcs/performance-tests/RTEvent/lib/Makefile: - Updated dependencies. + Updated dependencies. Thu Dec 11 15:38:08 2003 Jeff Parsons <j.parsons@vanderbilt.edu> @@ -85,29 +91,29 @@ Thu Dec 11 14:03:16 2003 Jeff Parsons <j.parsons@vanderbilt.edu> Thu Dec 11 14:01:34 2003 Balachandran Natarajan <bala@dre.vanderbilt.edu> - * docs/tutorials/Quoter/Simple/ImprovedServer/Quoter_Simple_ImprovedServer.mpc: - * examples/POA/DSI/POA_DSI.mpc: - * orbsvcs/examples/Notify/ThreadPool/Notify_ThreadPool.mpc: - * orbsvcs/performance-tests/RTEvent/lib/RTEC_Perf.mpc: - * orbsvcs/tests/Notify/performance-tests/Filter/Filter.mpc: - * orbsvcs/tests/Notify/performance-tests/Throughput/Throughput.mpc: - * tao/DynamicInterface/DynamicInterface.mpc: - * tests/RTScheduling/Scheduling_Interceptor/Scheduler_Interceptor.mpc: + * docs/tutorials/Quoter/Simple/ImprovedServer/Quoter_Simple_ImprovedServer.mpc: + * examples/POA/DSI/POA_DSI.mpc: + * orbsvcs/examples/Notify/ThreadPool/Notify_ThreadPool.mpc: + * orbsvcs/performance-tests/RTEvent/lib/RTEC_Perf.mpc: + * orbsvcs/tests/Notify/performance-tests/Filter/Filter.mpc: + * orbsvcs/tests/Notify/performance-tests/Throughput/Throughput.mpc: + * tao/DynamicInterface/DynamicInterface.mpc: + * tests/RTScheduling/Scheduling_Interceptor/Scheduler_Interceptor.mpc: - Added avoids += minimum_corba. + Added avoids += minimum_corba. Thu Dec 11 19:30:33 UTC 2003 Johnny Willemsen <jwillemsen@remedy.nl> - * tao/RTScheduling/Makefile.bor: - Don't install idl files + * tao/RTScheduling/Makefile.bor: + Don't install idl files Thu Dec 11 13:50:41 2003 Chris Cleeland <cleeland_c@ociweb.com> - * orbsvcs/orbsvcs/PortableGroup/UIPMC_Transport.cpp - (handle_input): + * orbsvcs/orbsvcs/PortableGroup/UIPMC_Transport.cpp + (handle_input): - Corrected the class name on handle_input definition. That'll - teach me to blindly use merge tools. + Corrected the class name on handle_input definition. That'll + teach me to blindly use merge tools. Thu Dec 11 18:13:12 UTC 2003 Johnny Willemsen <jwillemsen@remedy.nl> @@ -142,45 +148,45 @@ Thu Dec 11 18:13:12 UTC 2003 Johnny Willemsen <jwillemsen@remedy.nl> Thu Dec 11 10:45:53 2003 Chris Cleeland <cleeland_c@ociweb.com> - * orbsvcs/orbsvcs/PortableGroup/UIPMC_Transport.cpp - (handle_input): + * orbsvcs/orbsvcs/PortableGroup/UIPMC_Transport.cpp + (handle_input): - Updated this to reflect the new APIs from the PMB changes. - Thanks to Ossama Othman <ossama@dre.vanderbilt.edu> for pointing - out my oversight. + Updated this to reflect the new APIs from the PMB changes. + Thanks to Ossama Othman <ossama@dre.vanderbilt.edu> for pointing + out my oversight. Thu Dec 11 07:49:01 2003 Balachandran Natarajan <bala@dre.vanderbilt.edu> - * tests/RTCORBA/Thread_Pool/RTCORBA_Thread_Pool.mpc: + * tests/RTCORBA/Thread_Pool/RTCORBA_Thread_Pool.mpc: - Fixed a typo. + Fixed a typo. Wed Dec 10 09:23:35 2003 Balachandran Natarajan <bala@dre.vanderbilt.edu> - * examples/POA/Adapter_Activator/POA_Adapter_Activator.mpc: - * examples/POA/DSI/POA_DSI.mpc: - * examples/POA/Default_Servant/POA_Default_Servant.mpc: - * examples/POA/Explicit_Activation/POA_Explicit_Activation.mpc: - * examples/POA/FindPOA/POA_FindPOA.mpc: - * examples/POA/Forwarding/POA_Forwarding.mpc: - * examples/POA/Generic_Servant/POA_Generic_Servant.mpc: - * examples/POA/Loader/POA_Loader.mpc: - * examples/POA/NewPOA/POA_NewPOA.mpc: - * examples/POA/On_Demand_Activation/POA_On_Demand_Activation.mpc: - * examples/POA/POA_BiDir/POA_BiDir.mpc: - * examples/POA/TIE/POA_TIE.mpc: - * orbsvcs/tests/Bug_1630_Regression/test.mpc: - * tests/AMI_Buffering/client.cpp: - * tests/Bug_1639_Regression/Makefile - * tests/Bug_1639_Regression/Makefile.Bug_1639_testclient: - * tests/POA/Identity/Identity.mpc - * tests/RTCORBA/Linear_Priority/RTCORBA_Linear_Priority.mpc: - * tests/RTCORBA/MT_Client_Protocol_Priority/RTCORBA_MT_Client_Protocol_Priority.mpc: - * tests/RTCORBA/Policies/Policies.mpc: - * tests/RTCORBA/Policy_Combinations/RTCORBA_Policy_Combinations.mpc: - * tests/RTCORBA/Private_Connection/RTCORBA_Private_Connection.mpc: - * tests/RTCORBA/Server_Protocol/RTCORBA_Server_Protocol.mpc: - * tests/RTCORBA/Thread_Pool/RTCORBA_Thread_Pool.mpc: + * examples/POA/Adapter_Activator/POA_Adapter_Activator.mpc: + * examples/POA/DSI/POA_DSI.mpc: + * examples/POA/Default_Servant/POA_Default_Servant.mpc: + * examples/POA/Explicit_Activation/POA_Explicit_Activation.mpc: + * examples/POA/FindPOA/POA_FindPOA.mpc: + * examples/POA/Forwarding/POA_Forwarding.mpc: + * examples/POA/Generic_Servant/POA_Generic_Servant.mpc: + * examples/POA/Loader/POA_Loader.mpc: + * examples/POA/NewPOA/POA_NewPOA.mpc: + * examples/POA/On_Demand_Activation/POA_On_Demand_Activation.mpc: + * examples/POA/POA_BiDir/POA_BiDir.mpc: + * examples/POA/TIE/POA_TIE.mpc: + * orbsvcs/tests/Bug_1630_Regression/test.mpc: + * tests/AMI_Buffering/client.cpp: + * tests/Bug_1639_Regression/Makefile + * tests/Bug_1639_Regression/Makefile.Bug_1639_testclient: + * tests/POA/Identity/Identity.mpc + * tests/RTCORBA/Linear_Priority/RTCORBA_Linear_Priority.mpc: + * tests/RTCORBA/MT_Client_Protocol_Priority/RTCORBA_MT_Client_Protocol_Priority.mpc: + * tests/RTCORBA/Policies/Policies.mpc: + * tests/RTCORBA/Policy_Combinations/RTCORBA_Policy_Combinations.mpc: + * tests/RTCORBA/Private_Connection/RTCORBA_Private_Connection.mpc: + * tests/RTCORBA/Server_Protocol/RTCORBA_Server_Protocol.mpc: + * tests/RTCORBA/Thread_Pool/RTCORBA_Thread_Pool.mpc: Wed Dec 10 16:58:08 2003 Jeff Parsons <j.parsons@vanderbilt.edu> @@ -190,74 +196,74 @@ Wed Dec 10 16:58:08 2003 Jeff Parsons <j.parsons@vanderbilt.edu> Wed Dec 10 16:39:25 2003 Ossama Othman <ossama@dre.vanderbilt.edu> - * tao/Synch_Invocation.cpp (handle_system_exception): + * tao/Synch_Invocation.cpp (handle_system_exception): - Corrected repository IDs for non-fatal system exceptions being - detected (TRANSIENT, OBJ_ADAPTER, NO_RESPONSE). The - "IDL:omg.org/CORBA/" prefix and ":1.0" suffix were missing, thus - preventing code that should have been run in the presence of - these exceptions from being run. + Corrected repository IDs for non-fatal system exceptions being + detected (TRANSIENT, OBJ_ADAPTER, NO_RESPONSE). The + "IDL:omg.org/CORBA/" prefix and ":1.0" suffix were missing, thus + preventing code that should have been run in the presence of + these exceptions from being run. - Attempt to try another profile in the non-fatal exception case. - This fixes a problem where valid profiles were not retried when - they should have been. + Attempt to try another profile in the non-fatal exception case. + This fixes a problem where valid profiles were not retried when + they should have been. - * orbsvcs/orbsvcs/LoadBalancing/LB_LeastLoaded.h - (LL_DEFAULT_LOAD_PERCENT_DIFF_CUTOFF): + * orbsvcs/orbsvcs/LoadBalancing/LB_LeastLoaded.h + (LL_DEFAULT_LOAD_PERCENT_DIFF_CUTOFF): - New constant used in thundering herd fix described below. + New constant used in thundering herd fix described below. - * orbsvcs/orbsvcs/LoadBalancing/LB_LeastLoaded.cpp (get_location): + * orbsvcs/orbsvcs/LoadBalancing/LB_LeastLoaded.cpp (get_location): - Fixed problem where least loaded object group member was not - always selected, thus reducing scalability. + Fixed problem where least loaded object group member was not + always selected, thus reducing scalability. - If two locations have loads that differ by a "small" amount - (i.e. LL_DEFAULT_LOAD_PERCENT_DIFF_CUTOFF), choose one of them - at random. This helps alleviate a thundering herd phenomenon - that could occur when an object group member becomes available - for request handling. Thanks to Carlos, Marina and Jody at ATD - for coming up with this solution to this form of the thundering - herd problem. + If two locations have loads that differ by a "small" amount + (i.e. LL_DEFAULT_LOAD_PERCENT_DIFF_CUTOFF), choose one of them + at random. This helps alleviate a thundering herd phenomenon + that could occur when an object group member becomes available + for request handling. Thanks to Carlos, Marina and Jody at ATD + for coming up with this solution to this form of the thundering + herd problem. - * orbsvcs/orbsvcs/LoadBalancing/LB_ServerRequestInterceptor.cpp - (receive_request_service_contexts): + * orbsvcs/orbsvcs/LoadBalancing/LB_ServerRequestInterceptor.cpp + (receive_request_service_contexts): - Always allow LoadMonitor and LoadAlert operations to proceed - regardless of the alert condition. Addresses a problem where it - was not possible to disable alert states and query loads from a - LoadMonitor collocated with a server once a load alert was - enabled. + Always allow LoadMonitor and LoadAlert operations to proceed + regardless of the alert condition. Addresses a problem where it + was not possible to disable alert states and query loads from a + LoadMonitor collocated with a server once a load alert was + enabled. - * orbsvcs/performance-tests/LoadBalancing/ORBInitializer.h: + * orbsvcs/performance-tests/LoadBalancing/ORBInitializer.h: - Improved documentation of some class members. + Improved documentation of some class members. - * orbsvcs/performance-tests/LoadBalancing/ORBInitializer.cpp: + * orbsvcs/performance-tests/LoadBalancing/ORBInitializer.cpp: - Cosmetic change. + Cosmetic change. - * orbsvcs/performance-tests/LoadBalancing/RPS_Monitor.cpp (loads): + * orbsvcs/performance-tests/LoadBalancing/RPS_Monitor.cpp (loads): - If the elapsed-time is zero, set the load to zero to prevent a - division-by-zero error. + If the elapsed-time is zero, set the load to zero to prevent a + division-by-zero error. - * orbsvcs/performance-tests/LoadBalancing/Roundtrip.cpp - (test_method): + * orbsvcs/performance-tests/LoadBalancing/Roundtrip.cpp + (test_method): - Commented out a bunch of code that should not be used when - performing benchmarks. + Commented out a bunch of code that should not be used when + performing benchmarks. Wed Dec 10 13:58:41 2003 Chris Cleeland <cleeland_c@ociweb.com> - * tao/default_client.cpp (parse_args): Corrected erroneous option - name printed when -ORBConnectStrategy option parsing hits an - error. + * tao/default_client.cpp (parse_args): Corrected erroneous option + name printed when -ORBConnectStrategy option parsing hits an + error. - * tao/IIOP_Transport.cpp (recv): Corrected method name printed in - debug message. + * tao/IIOP_Transport.cpp (recv): Corrected method name printed in + debug message. - * tao/GIOP_Message_Base.cpp: + * tao/GIOP_Message_Base.cpp: * tao/GIOP_Message_Base.h: * tao/GIOP_Message_Base.i: * tao/GIOP_Message_Generator_Parser_Impl.inl: @@ -273,254 +279,254 @@ Wed Dec 10 13:58:41 2003 Chris Cleeland <cleeland_c@ociweb.com> * tao/Transport.cpp: * tao/Transport.h: * tao/orbconf.h: - * tao/Strategies/DIOP_Transport.cpp - * tao/Strategies/SHMIOP_Transport.cpp - * tao/Strategies/SHMIOP_Transport.h - - Integrated aggregate fix to infamous Parse Magic Bytes (PMB) - problem. The fix was originally developed in OCI's repository - and migrated here. An overall description of the problem and - solution follow, and a detailed changelog that follows the - development of this solution follows this entry. - - The old way of handling input naively assumed that a header would - never be broken across multiple read()s from the socket. Back in - the old days, this assumption was okay, because TAO performed a - recv_n() specifically for the header, and thus insured that it - wouldn't get less than a full header. However, when it was - changed to use a single read, that no longer worked. - - Further complicating matters was the incoming message queue, and - the myriad of methods that manipulated elements on the message - queue. - - The PMB changes rework things in the following ways: - - 1. The incoming message queue now holds only completely-read - GIOP messages awaiting processing. Each GIOP message is held in - a heap-allocated ACE_Message_Block. When the message has been - processed, the processing code is responsible for calling - release on the queued message to return the memory to the heap. - - 2. Since GIOP, by definition, assumes an ordered byte stream, - there can be only one partially read, i.e., uncompleted, message; - this is now held explicitly in TAO_Transport::uncompleted_message_ - as a pointer to a heap-allocated ACE_Message_Block of the exact - correct size for the GIOP message. - - 3. States for reading from the transport have been explicitly - defined rather than implied from values of other data members. - - This change also correctly handles correct receipt of GIOP - fragments: - - a. In the case of a GIOP 1.2 fragment, each fragment is a complete - GIOP message. It is, however, unable to be processed until all - fragments have arrived. - - b. Unlike a TCP segment, a GIOP fragment is not stateless; all - GIOP 1.2 fragments for a specific request must be sent in-order. - However, with transport multiplexing, other messages--including - fragments for other requests--may interleave these fragments. - - c. By design, there is no way to know the total number of payload - bytes across all GIOP fragments until the final fragment is - received. + * tao/Strategies/DIOP_Transport.cpp + * tao/Strategies/SHMIOP_Transport.cpp + * tao/Strategies/SHMIOP_Transport.h + + Integrated aggregate fix to infamous Parse Magic Bytes (PMB) + problem. The fix was originally developed in OCI's repository + and migrated here. An overall description of the problem and + solution follow, and a detailed changelog that follows the + development of this solution follows this entry. + + The old way of handling input naively assumed that a header would + never be broken across multiple read()s from the socket. Back in + the old days, this assumption was okay, because TAO performed a + recv_n() specifically for the header, and thus insured that it + wouldn't get less than a full header. However, when it was + changed to use a single read, that no longer worked. + + Further complicating matters was the incoming message queue, and + the myriad of methods that manipulated elements on the message + queue. + + The PMB changes rework things in the following ways: + + 1. The incoming message queue now holds only completely-read + GIOP messages awaiting processing. Each GIOP message is held in + a heap-allocated ACE_Message_Block. When the message has been + processed, the processing code is responsible for calling + release on the queued message to return the memory to the heap. + + 2. Since GIOP, by definition, assumes an ordered byte stream, + there can be only one partially read, i.e., uncompleted, message; + this is now held explicitly in TAO_Transport::uncompleted_message_ + as a pointer to a heap-allocated ACE_Message_Block of the exact + correct size for the GIOP message. + + 3. States for reading from the transport have been explicitly + defined rather than implied from values of other data members. + + This change also correctly handles correct receipt of GIOP + fragments: + + a. In the case of a GIOP 1.2 fragment, each fragment is a complete + GIOP message. It is, however, unable to be processed until all + fragments have arrived. + + b. Unlike a TCP segment, a GIOP fragment is not stateless; all + GIOP 1.2 fragments for a specific request must be sent in-order. + However, with transport multiplexing, other messages--including + fragments for other requests--may interleave these fragments. + + c. By design, there is no way to know the total number of payload + bytes across all GIOP fragments until the final fragment is + received. ---------------------------------------- Thu Mar 20 12:34:24 2003 Chris Cleeland <cleeland_c@ociweb.com> - * all: First merge back to OCI. + * all: First merge back to OCI. Thu Mar 20 12:17:23 2003 Chris Cleeland <cleeland_c@ociweb.com> - * tao/Strategies/DIOP_Transport.cpp (handle_input_i): - * orbsvcs/orbsvcs/PortableGroup/UIPMC_Transport.cpp - (handle_input_i): + * tao/Strategies/DIOP_Transport.cpp (handle_input_i): + * orbsvcs/orbsvcs/PortableGroup/UIPMC_Transport.cpp + (handle_input_i): - Update to use new parsing and message processing methods - consistent with PMB changes. + Update to use new parsing and message processing methods + consistent with PMB changes. - * tao/Transport.cpp (process_queue_head): Corrected a possible memory - leak where, if process_parsed_messages returned with an error, - the TAO_Queued_Data instance wouldn't be released. + * tao/Transport.cpp (process_queue_head): Corrected a possible memory + leak where, if process_parsed_messages returned with an error, + the TAO_Queued_Data instance wouldn't be released. Tue Mar 18 14:57:07 2003 Chris Cleeland <cleeland_c@ociweb.com> - * VERSION: - * ChangeLog: - * tao/tao.mpc: - * tao/Version.h: - * tao/IIOP_Transport.cpp: - * tao/MCAST_Parser.cpp: - * tao/RTCORBA/Linear_Priority_Mapping.cpp: - * tao/RTCORBA/RT_ORBInitializer.cpp: - * tao/RTCORBA/RT_ORBInitializer.h: - * tao/RTCORBA/RT_ORB_Loader.cpp: - * tao/RTCORBA/RT_Protocols_Hooks.cpp: + * VERSION: + * ChangeLog: + * tao/tao.mpc: + * tao/Version.h: + * tao/IIOP_Transport.cpp: + * tao/MCAST_Parser.cpp: + * tao/RTCORBA/Linear_Priority_Mapping.cpp: + * tao/RTCORBA/RT_ORBInitializer.cpp: + * tao/RTCORBA/RT_ORBInitializer.h: + * tao/RTCORBA/RT_ORB_Loader.cpp: + * tao/RTCORBA/RT_Protocols_Hooks.cpp: - Merged in changes from TAO 1.3.1. The actual tag for the merge - was pmb_branch_mainline_mergeout_1. The command was + Merged in changes from TAO 1.3.1. The actual tag for the merge + was pmb_branch_mainline_mergeout_1. The command was - $ cvs update -jpmb_branch_start -jpmb_branch_mainline_mergeout_1 + $ cvs update -jpmb_branch_start -jpmb_branch_mainline_mergeout_1 - after manually determining which files needed updates. + after manually determining which files needed updates. Mon Mar 17 11:09:00 2003 Chris Cleeland <cleeland_c@ociweb.com> - * tao/GIOP_Message_Base.cpp (check_for_valid_header): Optimized - implementation. + * tao/GIOP_Message_Base.cpp (check_for_valid_header): Optimized + implementation. - * tao/GIOP_Message_Base.* (header_length): Moved to inline for - optimization. + * tao/GIOP_Message_Base.* (header_length): Moved to inline for + optimization. - * tao/GIOP_Message_State.cpp (TAO_Debug_Msg_Emitter_Guard): - Improved implementation so it didn't make copies of the message - even when it was never going to output anything due to debug - level being lower than the value that would cause debug output. + * tao/GIOP_Message_State.cpp (TAO_Debug_Msg_Emitter_Guard): + Improved implementation so it didn't make copies of the message + even when it was never going to output anything due to debug + level being lower than the value that would cause debug output. - * tao/GIOP_Message_State.cpp (take_values_from_message_block): - * tao/GIOP_Message_State.cpp (set_version_info_from_buffer): - * tao/GIOP_Message_State.cpp (set_byte_order_info_from_buffer): - * tao/GIOP_Message_State.cpp (set_payload_size_from_buffer): + * tao/GIOP_Message_State.cpp (take_values_from_message_block): + * tao/GIOP_Message_State.cpp (set_version_info_from_buffer): + * tao/GIOP_Message_State.cpp (set_byte_order_info_from_buffer): + * tao/GIOP_Message_State.cpp (set_payload_size_from_buffer): - Removed use of TAO_Debug_Msg_Emitter_Guard. + Removed use of TAO_Debug_Msg_Emitter_Guard. - * tao/GIOP_Message_State.inl (set_payload_size_from_buffer): + * tao/GIOP_Message_State.inl (set_payload_size_from_buffer): - Moved to inline file. + Moved to inline file. - * tao/Incoming_Message_Queue.h (TAO_Incoming_Message_Queue): - Changed name of member 'queued_data_' to 'last_added_' to be - more clear about what the pointer actually points to. + * tao/Incoming_Message_Queue.h (TAO_Incoming_Message_Queue): + Changed name of member 'queued_data_' to 'last_added_' to be + more clear about what the pointer actually points to. - Added documentation that describes the structure of the linked - list so that hopefully future maintainers don't have to go - through the same learning curve I did. + Added documentation that describes the structure of the linked + list so that hopefully future maintainers don't have to go + through the same learning curve I did. - * tao/Incoming_Message_Queue.*: Updates to reflect last_added_ - member name change. + * tao/Incoming_Message_Queue.*: Updates to reflect last_added_ + member name change. - * tao/Incoming_Message_Queue.cpp (dequeue_head): Add check to not - allow dequeuing when size is zero, plus we now zero-out the - last_added_ pointer after we've dequeued the last item in the - list. + * tao/Incoming_Message_Queue.cpp (dequeue_head): Add check to not + allow dequeuing when size is zero, plus we now zero-out the + last_added_ pointer after we've dequeued the last item in the + list. - * tao/Incoming_Message_Queue.cpp (clone_mb_nocopy_size): Fix - setting and copying of flags on the message and data blocks so - that the DONT_DELETE flag specifically doesn't get copied. This - cured a HUGE memory leak that was causing substantial - performance problems. + * tao/Incoming_Message_Queue.cpp (clone_mb_nocopy_size): Fix + setting and copying of flags on the message and data blocks so + that the DONT_DELETE flag specifically doesn't get copied. This + cured a HUGE memory leak that was causing substantial + performance problems. - * tao/Incoming_Message_Queue.cpp (make_uncompleted_message): - * tao/Incoming_Message_Queue.cpp (make_completed_message): + * tao/Incoming_Message_Queue.cpp (make_uncompleted_message): + * tao/Incoming_Message_Queue.cpp (make_completed_message): - Cache values used over and over again within these methods such - as header and message block lengths + Cache values used over and over again within these methods such + as header and message block lengths - * tao/Transport.cpp (handle_input_i): + * tao/Transport.cpp (handle_input_i): - Added flag noting that we enqueued a message. + Added flag noting that we enqueued a message. - Changed how re-reading is performed; see entry for - try_to_complete for information. + Changed how re-reading is performed; see entry for + try_to_complete for information. - Used ACE_CDR::grow to grow the uncompleted_message message block - once the payload size is known. This insures that the growth - happens with proper alignment constraints. + Used ACE_CDR::grow to grow the uncompleted_message message block + once the payload size is known. This insures that the growth + happens with proper alignment constraints. - * tao/Transport.* (try_to_complete): Added this method which - tries to complete whatever partial message is held in - uncompleted_message_. The difference between doing this and - simply revisiting the top of handle_input_i is that this will - only try to read enough to complete the message rather than also - read another partial. + * tao/Transport.* (try_to_complete): Added this method which + tries to complete whatever partial message is held in + uncompleted_message_. The difference between doing this and + simply revisiting the top of handle_input_i is that this will + only try to read enough to complete the message rather than also + read another partial. - * tao/Transport.cpp (process_queue_head): Added check to insure - that we don't try to do any processing when there's nothing in - the queue. Also, removed old code leftover from when the - incoming queue could have partial GIOP messages in it. Finally, - insured that the return value matched previous requirements. + * tao/Transport.cpp (process_queue_head): Added check to insure + that we don't try to do any processing when there's nothing in + the queue. Also, removed old code leftover from when the + incoming queue could have partial GIOP messages in it. Finally, + insured that the return value matched previous requirements. Thu Mar 6 15:27:14 2003 Chris Cleeland <cleeland_c@ociweb.com> - * tao/Incoming_Message_Queue.cpp: Removed Shattering Encapsulation - hack around ace/Message_Block.h inclusion to make this link - properly on VC6. + * tao/Incoming_Message_Queue.cpp: Removed Shattering Encapsulation + hack around ace/Message_Block.h inclusion to make this link + properly on VC6. - * tao/GIOP_Message_State.cpp: Moved initialization of - TAO_Debug_Msg_Emitter_Guard::MAGIC_LENGTH outside the class - declaration. + * tao/GIOP_Message_State.cpp: Moved initialization of + TAO_Debug_Msg_Emitter_Guard::MAGIC_LENGTH outside the class + declaration. Wed Mar 5 13:08:37 2003 Chris Cleeland <cleeland_c@ociweb.com> - * tao/GIOP_Message_Generator_Parser_Impl.inl (check_revision): - Modified to only explicitly allow versions of GIOP that TAO - supports, and no others. Note that this change means that - whenever a new version of GIOP gets added to TAO, this function - must be updated, too. Hopefully that won't happen so often that - this becomes horribly burdensome. Thanks to Chad Elliott and - Paul Calabrese for ideas in getting this right. + * tao/GIOP_Message_Generator_Parser_Impl.inl (check_revision): + Modified to only explicitly allow versions of GIOP that TAO + supports, and no others. Note that this change means that + whenever a new version of GIOP gets added to TAO, this function + must be updated, too. Hopefully that won't happen so often that + this becomes horribly burdensome. Thanks to Chad Elliott and + Paul Calabrese for ideas in getting this right. - * tao/Incoming_Message_Queue.* (make_uncompleted_message): + * tao/Incoming_Message_Queue.* (make_uncompleted_message): - This now behaves similar to make_completed_message, i.e., it - allocates the message block held in TAO_Queued_Data and advnaced - the rd_ptr past the bytes it consumes. + This now behaves similar to make_completed_message, i.e., it + allocates the message block held in TAO_Queued_Data and advnaced + the rd_ptr past the bytes it consumes. - * tao/Incoming_Message_Queue.cpp: Eliminated hacks on - ACE_Message_Block and replaced with functions - clone_mb_nocopy_size() and copy_mb_span() that utilize ONLY the - public API. make_completed_message() and - make_uncompleted_message() now use these. + * tao/Incoming_Message_Queue.cpp: Eliminated hacks on + ACE_Message_Block and replaced with functions + clone_mb_nocopy_size() and copy_mb_span() that utilize ONLY the + public API. make_completed_message() and + make_uncompleted_message() now use these. - * tao/Transport.cpp (handle_input_i): Enlarged the buffer size and - replaced the magic number for the maximum number of re-read - attempts with the manifest constant - TAO_MAX_TRANSPORT_REREAD_ATTEMPTS, now defined in orbconf.h. + * tao/Transport.cpp (handle_input_i): Enlarged the buffer size and + replaced the magic number for the maximum number of re-read + attempts with the manifest constant + TAO_MAX_TRANSPORT_REREAD_ATTEMPTS, now defined in orbconf.h. - Changed uses of make_uncompleted_message() to be consistent with - changes outlined above. + Changed uses of make_uncompleted_message() to be consistent with + changes outlined above. - Corrected a bug where, upon re-reading after completely emptying - the stack-allocated message block, I forgot to reset the - rd_ptr/wr_ptr in the message block. + Corrected a bug where, upon re-reading after completely emptying + the stack-allocated message block, I forgot to reset the + rd_ptr/wr_ptr in the message block. - * tao/orbconf.h: Added new manifest constant - TAO_MAX_TRANSPORT_REREAD_ATTEMPTS. Documentation is in the file. + * tao/orbconf.h: Added new manifest constant + TAO_MAX_TRANSPORT_REREAD_ATTEMPTS. Documentation is in the file. Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> - * TAO/tao/GIOP_Message_Base.cpp: + * TAO/tao/GIOP_Message_Base.cpp: - Eliminate parse_incoming_messages(), missing_data(), - extract_next_message(), consolidate_node(), - consolidate_fragments(), get_message_data(), make_queued_data(). + Eliminate parse_incoming_messages(), missing_data(), + extract_next_message(), consolidate_node(), + consolidate_fragments(), get_message_data(), make_queued_data(). - Add new methods set_queued_data_from_message_header() and - check_for_valid_header(). + Add new methods set_queued_data_from_message_header() and + check_for_valid_header(). - * TAO/tao/GIOP_Message_Base.h: + * TAO/tao/GIOP_Message_Base.h: - Eliminate #if0 parse_incoming_messages(), missing_data(), - extract_next_message(), consolidate_node(), - consolidate_fragments(), get_message_data(), and - make_queued_data() declarations. + Eliminate #if0 parse_incoming_messages(), missing_data(), + extract_next_message(), consolidate_node(), + consolidate_fragments(), get_message_data(), and + make_queued_data() declarations. - Add declarations for set_queued_data_from_message_header() and - check_for_valid_header(). + Add declarations for set_queued_data_from_message_header() and + check_for_valid_header(). - Make message_type() static. + Make message_type() static. - * TAO/tao/GIOP_Message_Generator_Parser_Impl.inl: + * TAO/tao/GIOP_Message_Generator_Parser_Impl.inl: - Reimplement check_revision() to be exact about which GIOP - revisions are okay. This is different from before where it - loosely assumed that everything was valid unless it want higher - than a particular range. + Reimplement check_revision() to be exact about which GIOP + revisions are okay. This is different from before where it + loosely assumed that everything was valid unless it want higher + than a particular range. - * TAO/tao/GIOP_Message_Lite.cpp: + * TAO/tao/GIOP_Message_Lite.cpp: Remove parse_incoming_messages() thru consolidate_fragments(); remove make_queued_data(). @@ -528,7 +534,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Add set_queued_data_from_message_header() and check_for_valid_header(). - * TAO/tao/GIOP_Message_Lite.h: + * TAO/tao/GIOP_Message_Lite.h: Remove declarations for parse_incoming_messages() thru consolidate_fragments(); remove make_queued_data(). @@ -536,7 +542,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Add declarations for set_queued_data_from_message_header() and check_for_valid_header(). - * TAO/tao/GIOP_Message_State.cpp: + * TAO/tao/GIOP_Message_State.cpp: Add TAO_Debug_Message_Emitter_Guard class within the compilation unit. @@ -561,7 +567,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Change argument type for read_ulong() to 'const char*'. - * TAO/tao/GIOP_Message_State.h: + * TAO/tao/GIOP_Message_State.h: Add default value for CTOR arguments so we don't have to pass them. @@ -583,12 +589,12 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> #if0 base_ member variable. - * TAO/tao/GIOP_Message_State.inl: + * TAO/tao/GIOP_Message_State.inl: Add definitions for giop_version(), more_fragments(), and message_type(). - * TAO/tao/Incoming_Message_Queue.cpp: + * TAO/tao/Incoming_Message_Queue.cpp: Add #include for ace/Message_Block.h, but shatter encapsulation and effectively extend its interface so that we can clone spans of @@ -596,7 +602,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Modify copy_tail() to use new TAO_Queued_Data member names. - * TAO/tao/Incoming_Message_Queue.cpp (class TAO_Queued_Data): + * TAO/tao/Incoming_Message_Queue.cpp (class TAO_Queued_Data): Update CTORs to initialize new this->current_state_ data member. @@ -605,7 +611,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Rename get_queued_data() to make_queued_data(). - * TAO/tao/Incoming_Message_Queue.h: + * TAO/tao/Incoming_Message_Queue.h: Changed #include <Pluggable_Messaging_Utils.h> to #include <Pluggable_Messaging.h> @@ -615,7 +621,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> #if0 get_node(). - * TAO/tao/Incoming_Message_Queue.h (class TAO_Queued_Data): + * TAO/tao/Incoming_Message_Queue.h (class TAO_Queued_Data): Make CTORs protected so creation can only go through new factory methods, i.e., make_completed_message() and @@ -634,7 +640,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Documentation for this->byte_order_. - * TAO/tao/Incoming_Message_Queue.inl: + * TAO/tao/Incoming_Message_Queue.inl: Replace uses of this->missing_data_ with this->missing_data_bytes_. @@ -643,14 +649,14 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Add release() method. - * TAO/tao/Pluggable_Messaging.h: + * TAO/tao/Pluggable_Messaging.h: #if0 parse_incoming_messages() through consolidate_fragments(). Add declarations for check_for_valid_header() and set_queued_data_from_message_header(). - * TAO/tao/Transport.cpp: + * TAO/tao/Transport.cpp: Add #include <ace/Min_Max.h> @@ -673,7 +679,7 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> consolidate_message_queue(), consolidate_extra_messages(), and make_queued_data(). - * TAO/tao/Transport.h: + * TAO/tao/Transport.h: #if0 declarations for parse_consolidate_messages(), parse_incoming_messages(), missing_data(), consolidate_message(), consolidate_fragments(), @@ -682,31 +688,31 @@ Wed Feb 26 21:44:34 2003 Chris Cleeland <cleeland_c@ociweb.com> Add this->uncompleted_message_ data member. - * TAO/tao/Strategies/DIOP_Transport.cpp: + * TAO/tao/Strategies/DIOP_Transport.cpp: Change 1st arg of this->recv() in handle_input_i() to wr_ptr() from rd_ptr(). - * TAO/tao/Strategies/SHMIOP_Transport.cpp: + * TAO/tao/Strategies/SHMIOP_Transport.cpp: #if0 consolidate_message(). - * TAO/tao/Strategies/SHMIOP_Transport.h: + * TAO/tao/Strategies/SHMIOP_Transport.h: #if0 declaration for consolidate_message(). - * TAO/tests/AMI/run_test.pl: + * TAO/tests/AMI/run_test.pl: Increase default number of iterations from 1 to 10000. Earlier versions of the PMB changes succeeded just fine with low numbers of iterations, but began to fail miserably when the number of iterations climbed. - * TAO/tests/BiDirectional/run_test.pl: + * TAO/tests/BiDirectional/run_test.pl: Added level 10 debug. - * TAO/tests/InterOp-Naming/INS_test_client.cpp: + * TAO/tests/InterOp-Naming/INS_test_client.cpp: Changes that don't appear to be related in any way to PMB... @@ -952,10 +958,10 @@ Mon Dec 8 19:33:03 2003 Balachandran Natarajan <bala@dre.vanderbilt.edu> * tao/Synch_Invocation.cpp: - Fixed a minor problem with reinvocation. We were not handling - errors properly, more precisely the restart errors if they show - up during wait_for_reply (). Thanks to Milan Cvetkovic for - motivating this change. + Fixed a minor problem with reinvocation. We were not handling + errors properly, more precisely the restart errors if they show + up during wait_for_reply (). Thanks to Milan Cvetkovic for + motivating this change. * tao/Messaging/Asynch_Invocation.cpp: diff --git a/TAO/orbsvcs/orbsvcs/PortableGroup/PG_Location_Hash.h b/TAO/orbsvcs/orbsvcs/PortableGroup/PG_Location_Hash.h index a1d25a88959..105ef25e6db 100644 --- a/TAO/orbsvcs/orbsvcs/PortableGroup/PG_Location_Hash.h +++ b/TAO/orbsvcs/orbsvcs/PortableGroup/PG_Location_Hash.h @@ -22,6 +22,7 @@ # pragma once #endif /* ACE_LACKS_PRAGMA_ONCE */ +#include "portablegroup_export.h" #include "orbsvcs/PortableGroupC.h" #include "ace/ACE.h" |